CityCyclingEdinburgh was launched on the 27th of October 2009 as "an experiment".
IT’S TRUE!
CCE is 16years old!
Well done to ALL posters
It soon became useful and entertaining. There are regular posters, people who add useful info occasionally and plenty more who drop by to watch. That's fine. If you want to add news/comments it's easy to register and become a member.
RULES No personal insults. No swearing.
when is a survey not a survey?
http://www.sustrans.org.uk/news/50-children-scotland-are-getting-school-foot-bike-or-scooter
am I being too cynical? if participation in the 'survey' is self-selecting, then its not really a 'survey' is it?
Just seems likely that those schools with higher rates are more likely to participate. in particular, i';d like some data on fee-paying schools.
All surveys are, to a greater or lesser extent, self-selecting. Apart from the census where participation is compulsory. So the question is how much self-selection is there (what's the response rate) and does this systematically bias the results? If the response rate is high, the scope for bias is reduced and vice versa.
In this, if it's biased, it would be relatively easy to fix (or patch) either by imputation or weighting. Imputation would be better - matching non-participating schools with non-participating schools on variables that are strongly related to travel to school and donating data from one to the other. This works very well, provided the modeling is good. My colleague has done it for various big Government surveys both internally - imputating data across cases from the same survey - and externally - imputing data in one survey using data from another.
Of the 2,569 schools in Scotland 2,017 (79%) took part, which is a pretty good sample.
The sampling rate for Scottish school children was a little lower - 69% by my calculation, possibly because the survey has a greater proportion of primary schools participating than secondary schools?
Yet given the size of the sample, it should be amenable to analysis by those skilled in such matters (not me!).
Robert
From SRD's link, a few more links leads to spreadsheet of National Results.
About a third of Councils have 100% of their primary schools replying. Out of 45 "Independent"/Private schools, 13 replied.
My quick analysis - change over time - 2008-2013 seems more relevant than this year's rates.
Rates of schoolchildren walking to school in ind. sector decreased fro, 27.2 in 2008 to 15.3 in 2013
Primary school walking rates decrease since 2008 (52-46); cycle rates up a little (3.4-5)
Scootering only obvious success for primary - from 1%-4.3%.
However, as the number of schools included has grown a lot, it could be that original participants were skewed to the more 'active' schools, and the decrease more reflects change in sample, than actual decrease.
Also notable is the driving of kids to nursery. Massive! I'm assuming this is because there is no obligation to attend a local nursery, whereas primaries are often 'full' and don't accept (many) out of catchment requests. Also 3-4 year olds less independent, but most can walk...
Independent school driving is big, perhaps reflects a combination of distance travelled and parental incomes...
LAs with surprisingly high driving rates include Dundee City and Glasgow City. Or maybe not that surprising for anyone who has spent time in either. Edinburgh City relatively low driving rate, but stubbornly refuses to decrease by much, mainly switching to 'park and stride' which may avoid congestion outside school gates but is still driving.
Cycling seems to be taking modal share from walking rather than motorised modes...
OTOH the response rates for urban areas seem pretty poor so it's hared to say how reliable the figures are.
You must log in to post.
Cycling in Edinburgh Flickr group
Video embedded using Easy Video Embed plugin
posts